Estimating Per-pixel Classification Confidence of Remote Sensing Images

Estimating Per-pixel Classification Confidence of Remote Sensing Images

Bibliographic Details
Main Author: Jiang, Shiguo
Language:English
Published: The Ohio State University / OhioLINK 2012
Subjects:
Geographic Information Science
Geography
Remote Sensing
spatial data quality
GIS
remote sensing
image classification
classification confidence
sample design
classification error
posterior probability
entropy
maximum likelihood
support vector machine
neural network
boosted decision tree
Online Access:http://rave.ohiolink.edu/etdc/view?acc_num=osu1354557859
id ndltd-OhioLink-oai-etd.ohiolink.edu-osu1354557859
record_format oai_dc
spelling ndltd-OhioLink-oai-etd.ohiolink.edu-osu13545578592021-08-03T06:06:46Z Estimating Per-pixel Classification Confidence of Remote Sensing Images Jiang, Shiguo Geographic Information Science Geography Remote Sensing spatial data quality GIS remote sensing image classification classification confidence sample design classification error posterior probability entropy maximum likelihood support vector machine neural network boosted decision tree Spatial data quality is an important topic in geographic information sciences and remote sensing. It has drawn attention from academic community, government agencies, and industry. Although great progress has been made on the spatial quality of interval and ratio data, the spatial uncertainty of nominal and ordinal data remains problematic. Land use land cover is one of the most important nominal data, which has broad impacts on our environment. The significance of Land use land cover change (LULCC) as an environmental factor calls for studies on the spatial data quality in LULCC. Remote sensing image classification is the most common source for LULCC. Therefore, the accuracy of remote sensing image classification is especially important. This dissertation aims to address the challenge to reporting classification confidence at pixel level. First, it provides a comprehensive literature review on previous studies. Then a rigorous evaluation of the main current methods is presented. Based on the literature review and evaluation, a new method is presented. The results are validated with complete coverage reference data. The estimated classification confidence is comparable across classifiers and thus indicate the performance of different classifiers. 2012-12-19 English text The Ohio State University / OhioLINK http://rave.ohiolink.edu/etdc/view?acc_num=osu1354557859 http://rave.ohiolink.edu/etdc/view?acc_num=osu1354557859 unrestricted This thesis or dissertation is protected by copyright: all rights reserved. It may not be copied or redistributed beyond the terms of applicable copyright laws.
collection NDLTD
language English
sources NDLTD
topic Geographic Information Science
Geography
Remote Sensing
spatial data quality
GIS
remote sensing
image classification
classification confidence
sample design
classification error
posterior probability
entropy
maximum likelihood
support vector machine
neural network
boosted decision tree
spellingShingle Geographic Information Science
Geography
Remote Sensing
spatial data quality
GIS
remote sensing
image classification
classification confidence
sample design
classification error
posterior probability
entropy
maximum likelihood
support vector machine
neural network
boosted decision tree
Jiang, Shiguo
Estimating Per-pixel Classification Confidence of Remote Sensing Images
author Jiang, Shiguo
author_facet Jiang, Shiguo
author_sort Jiang, Shiguo
title Estimating Per-pixel Classification Confidence of Remote Sensing Images
title_short Estimating Per-pixel Classification Confidence of Remote Sensing Images
title_full Estimating Per-pixel Classification Confidence of Remote Sensing Images
title_fullStr Estimating Per-pixel Classification Confidence of Remote Sensing Images
title_full_unstemmed Estimating Per-pixel Classification Confidence of Remote Sensing Images
title_sort estimating per-pixel classification confidence of remote sensing images
publisher The Ohio State University / OhioLINK
publishDate 2012
url http://rave.ohiolink.edu/etdc/view?acc_num=osu1354557859
work_keys_str_mv AT jiangshiguo estimatingperpixelclassificationconfidenceofremotesensingimages
_version_ 1719431102325587968

Similar Items